Manuscript preparation instructions

Figure 3 PSD of 200Hz 1/3 Octave 10-minute indoor data sampled at 10Hz

---

IoA AM Conference Cardiff 20 March 2014 
4 
Submitted 5 February 2014 
Figure 4 PSD of 250Hz 1/3 Octave 10-minute indoor data sampled at 10Hz 
The peaks in the PSD charts clearly show the blade pass frequency of the two turbines around 
0.8 Hz. The amplitude of the PSD charts is in units of dB (p-p re 20µPa) squared per Hz. 
It is possible to use the amplitude of the PSD spectrum peak to quantify an amplitude 
modulation value. This is an approach used by RenewableUK in their recently released AM 
assessment tool ‘OAM’. Comparison of the peak PSD result with the average PSD result may 
also be of quantitative value because it could be the peak AM that cause complaints. The 
peak spectrum (red) and average spectrum (blue) are shown in Figures 3 and 4.
OTHER AM CHARACTERISTICS 
It is unknown if the envelope of the modulated sound is more or less annoying than a constant 
AM (as seen in the example of figure 2 for the two asynchronous wind turbines at Leonards 
Hill using Fast response). 
Another unknown potential nuisance factor is the change in swish repetition rate. Figure 5 
shows 100ms LAeq data from a 73s sample outdoor DUO measurement at Leonards Hill.
Over approximately one-minute the observed swish repetition changes from twice blade pass 
frequency (BPF) to BPF. Closer inspection of the data shows that in the earlier part of the 
chart, when the blade pass frequencies are out of phase, there is only one data point 
representing each of the peaks. In this situation a 10Hz data rate can underestimate real peak 
values, even though the frequency of AM can be determined from spectrum analysis.
TIME CONSTANT CHOICE 
Fast response corresponds to a 125 ms time constant, Slow corresponds to a 1 second time 
constant and Impulse has a time constant of 35 ms. One must question if the Fast response of 
125ms is suitable for accurately tracking AM pressure changes caused by multiple turbines. 
Although now out of favour in IEC 61672, the Impulse response will produce a better peak to 
trough estimation for repetitive swish sounds, such as those shown at the start of the trace in 
Figure 5 (2 Hz repetition). However, even the Impulse response can underestimate real AM 
levels since the design goal from IEC 61672 for the relative response of A-frequency-
weighted, I-time-weighted sound level to a sequence of 5 ms, 4 kHz tonebursts is -8.8 dB for 
a 2 Hz repetition rate. 

---

IoA AM Conference Cardiff 20 March 2014 
5 
Submitted 5 February 2014 
HAAS EFFECT 
A situation will arise with multiple turbine blade swish AM where the time between 
successive swish observations is such that the time reduces to less than about 40ms. In this 
circumstance an increased perception of loudness will occur even though the AM levels 
remain relatively constant. Furthermore, an observer can be totally confused about the 
direction from which the sounds originate.